Awarri
Nigeria's open multilingual LLM, capped at 1,000 users free
Overview
Awarri is a Lagos AI and robotics company that built N-ATLAS with Nigeria's government — an open Llama-3-8B model fine-tuned on 400M+ tokens for Yoruba, Hausa, Igbo and Pidgin — plus LangEasy and RootLens, its crowdsourced data platforms for African languages and visual data.
Pricing
Pricing shown for reference only. These figures reflect RECATOOLS research as of 12 Jul 2026 and may be out of date or incomplete. This is not financial or purchasing advice — always confirm the current price on the provider’s official website before making any decision.
What you can produce with Awarri
- N-ATLAS: open Llama-3-8B model fine-tuned for Yoruba, Hausa, Igbo, Pidgin
- Published model weights on Hugging Face (NCAIR1/N-ATLaS)
- LangEasy crowdsourced language-data collection platform
- RootLens visual-data labeling and management platform
- Free use license for projects under 1,000 active users
- Commercial licensing path for larger deployments
ASEAN Perspective
Awarri in Southeast Asia
ASEAN-region availability and pricing notes coming soon. Drop the editorial team a note via /contact/ if you can supply local context (Singapore/Malaysia/Indonesia/Thailand/Vietnam).
Awarri's real contribution isn't a frontier model — N-ATLAS is an 8-billion-parameter Llama-3 derivative, nowhere near GPT- or Gemini-class capability — it's the data pipeline behind it. LangEasy crowdsources Yoruba, Hausa, Igbo, Pidgin and Ibibio text and speech at national scale, backed by the Federal Ministry of Communications and NITDA, and Awarri says it has trained and hired over 4,200 data operators toward a stated goal of 5,000 jobs by 2026. That's a genuinely rare thing: a for-profit company building African-language AI infrastructure with government money behind it rather than as a side project. The catch is the license — free use caps out at 1,000 active end-users before you need a commercial deal directly with Awarri, so this isn't a drop-in open-source model for a product with real scale. For researchers, NGOs, and small teams building Nigerian-language tools, it's one of the few open options; for anything commercial-scale, budget for the paid license.
What people say
Awarri is a Lagos-based AI and robotics company positioned as Nigeria's answer to the frontier-model race, though its actual output so far is smaller and more infrastructure-focused than that framing suggests. Its highest-profile release is N-ATLAS (also styled N-ATLaS), launched in 2026 as an open-source multilingual, multimodal model fine-tuned from Meta's Llama-3 8B on more than 400 million tokens of instruction data covering Yoruba, Hausa, Igbo, Nigerian Pidgin and English. It was built jointly with Nigeria's Federal Ministry of Communications, Innovation and Digital Economy as part of the Nigerian Languages AI Initiative, and the model weights are published on Hugging Face under NCAIR1/N-ATLaS.
The license is worth reading closely: it's modeled on Apache 2.0/MIT but caps free use at organizations, institutions or projects with no more than 1,000 active end-users — beyond that, you need a commercial license from Awarri and the ministry. That structure suggests the model is aimed more at researchers, students, and small civic-tech projects than at companies trying to ship a consumer product at scale.
Behind N-ATLAS sits Awarri's actual moat: LangEasy, a crowdsourced data-collection platform where Nigerians contribute voice and text samples in their own languages, and RootLens, a visual-data labeling and management platform. The company says these have trained and employed more than 4,200 data operators and engineers so far, working toward a stated goal of 5,000 hired by 2026 — a jobs-and-data-sovereignty pitch as much as a product one. Awarri also ran a 2025 Developer Challenge to get local builders shipping on top of the model, covering use cases in education, governance, health and agriculture.
There's no independent review coverage (G2, Capterra, Trustpilot) to draw on here — Awarri is infrastructure, not a SaaS product with a user base leaving star ratings, and its footprint outside developer and NGO circles is minimal. What's verifiable is the model itself (published weights, benchmarked against Llama-3 8B baselines), the government partnership, and the license terms; what's not independently verifiable is how many organizations have actually shipped production use of N-ATLAS.
Summary of public user & expert reviews, compiled by RECATOOLS.
About this listing
This entry was compiled from publicly available data including Awarri's official website, press releases, documentation, and reputable third-party publications. RECATOOLS is not affiliated with Awarri unless explicitly stated.
Third-party AI tools update their pricing, features, availability, and policies frequently. Information here may be outdated by the time you read this — we make reasonable efforts to keep listings current, but cannot guarantee absolute accuracy.
For the latest details, please refer to Awarri directly →
Spotted something out of date? Suggest an update →
Awarri in the news
Alternatives to Awarri
More in LLMs & Chat
